Professional Documents
Culture Documents
DOI: 10.5923/c.jmea.201502.10
Abstract Low speed wind tunnel testing has evolved over the last century and has become a cornerstone in the
development of aviation vehicles. Study of the effects of aerodynamics and its influence are substantial in the design of
aircrafts and much fluid effect based engineering machines. The forces involved in the dynamic interactions of fluids and
solids have only begun to be lifted, therefore the study of wind behavior is necessary to design large Airplanes. This paper
provides an overview of the design parameters of a low speed wind tunnel and then the study of coefficient of lift (CL) and
coefficient of drag (CD) for different length to diameter (L/D) ratios of a NACA 4412 aerofoil section shaped as per the
NACA standards of Aerofoil design. It was observed through flow visualization test that flow remained unseparated
throughout the test section for 70% of its characteristic dimension i.e. the breadth or height of the test section in this case. A
conditioning circuit was also developed in order to record the (CL) and (CD) values which gave smooth output data values.
The LM324 IC was used for noise and temperature output for good stability for dynamic data.
Keywords Coefficient of lift, Coefficient of drag, L/D ratio, Reynolds number, Laminar flow
force and motion are related by Newtons second law, and (3)
1. Introduction energy exchanges are governed by First Law of
Thermodynamics. In addition to these three principles,
Low speed wind tunnel testing has evolved over the last certain fluid properties and their variations with pressure and
century and has become a cornerstone in the development of temperature must be described mathematically with
aviation vehicles. Late in the 1970s, multiple groups of sufficient accuracy for each application. The objective of this
experts predicted that the need for aerodynamic experiments, paper is to study the aerodynamic aspects of designing a
particularly in the subsonic region of airflow, would rapidly subsonic Wind Tunnel and to study the existing aerofoil
disappear as computational fluid dynamics (CFD) would design to understand what parameters affect the lift and drag
take over. Though the multiple formidable improvements in coefficients. The available literatures that deal with the
CFD prevail today and computer power continues to evolve, design of wind tunnel or those components are illustrated
our mathematical models are still far for being a perfect firstly. Bradshaw and Mehta, 1979, presented design guide
analysis of an aircraft. The shapes of airplanes are lines for the main components of a wind tunnel, the fan, wide
determined by considerations of aerodynamics with varying angle diffuser, settling chamber, contraction, screens, honey
degrees of attention to performance, agility, stealth, combs and exit diffuser for small low speed wind tunnels.
procurement cost, operational cost, time to delivery and Milan Valajinac, 1970 summarized the design, construction
other aspect the customer may require for intended missions. and calibration of a subsonic wind tunnel. Raikan 2001, a
A primary tool of experimental aerodynamics is the wind subsonic wind tunnel design of closed circuit type for a range
tunnel. The productive and proper use of experimental of Mach No. in the test section up to 0.7 has been carried out
investigations in general wind tunnels in particular requires and a mathematical models describe the flow in each
applications of aerodynamic theory and computational component of the wind tunnel are made. The comparisons
methods in the planning off facilities, planning of between the results obtained from their design and the
experiments and the interpretation of resulting data. The literature were done and they showed that a good agreement
fundamental principles from which the equations used to exist with these results. NACA 4412 aerofoil is used to study
model low-speed aerodynamic flows are derived and are the parameters of design and flow parameters with reference
only three in number. These are (1) Mass is conserved, (2) to lift and drag. J. Colman 2010 investigated an airfoil, with
and without the Gurney mini-flaps, immersed in a low
* Corresponding author:
alisterdsouza14@gmail.com (Alister Gleason Dsouza)
Reynolds number turbulent flow. Lift and drag coefficients
Published online at http://journal.sapub.org/jmea were calculated for the plain wing and for the wing with
Copyright 2015 Scientific & Academic Publishing. All Rights Reserved mini-flaps of 1%, 1.5%, 2% and 2.5% height of the chord
48 Alister Gleason Dsouza et al.: Study of Aerofoil Design Parameters for Low Speed Wind Tunnel
and plotted as a function of the angle of attack. The the exclusion of characteristic length). Much low-speed
experimental data, including the power density spectrum of testing involves aircraft takeoff and landing configurations
the instantaneous longitudinal and vertical velocities and where the Mach number is typically in the 0.15-0.3 range. A
load, show that the Gurney mini-flap acts enhancing the lift primary decision is the choice of the minimum acceptable
coefficient of the airfoil coupled with an increased drag, value of Reynolds number. Because much of the low-speed
primarily due to the particular mini-flaps wake structure. testing is at high lift conditions, the effect of Reynolds
number on aerofoils at high lift must be considered.
Maximum lift and lift curve shape near stall for single
2. Wind Tunnel Design element airfoils vary considerably with Reynolds number up
to at least one million. For multi-element airfoils this range is
The first step in the design of a wind tunnel is to determine much greater. If the Mach number is taken as 0.2, the tunnel
the size and shape of the test section based on the intended velocity is about 67m/s. For this speed the unit Reynolds
uses of the facility. A major part of the following discussions number is a little less than 458,000m-1.
address tunnels for which the primary use is vehicle and
vehicle component testing, since many tunnels are built for ii) Open or Closed Return
this purpose. It is also assumed that a major part of the testing Another basic design consideration is whether the tunnel
will be force testing, where information is sought for will be of a return or non-return type. Almost all of the small
performance, fuel efficiency, stability or control of a vehicle research tunnels are of the non-return type, usually because
that may be an aircraft, an automobile, a submarine or of the lower construction costs. Power consumption for such
possibly others. The cross sectional area of the test section tunnels is usually not a significant factor in overall cost.
basically determines the overall size of the facility. The test Although there exist some larger tunnels of non-return
section size, speed and design will determine the required design, most of the large tunnels are of the return type the
power. The size of the facility will be the primary factor in majority being single return.
determining the structural or shell costs, and the power and
operating hours will determine the energy portion of the 2.1. Wind Tunnel Component Energy Losses
operational cost. The overall aerodynamic objective for most The case of wind tunnel flow is considered to be the case
wind tunnels is to obtain a flow in the test section that is near of flow through pipes. Hence, there are multiple losses
as possible to a parallel steady flow with uniform speed occurring due to friction at the boundaries. This resulting
throughout the test section. Perfection is not possible so loss of pressure is termed as pressure loss and it is the
every design is bounded by constraints that involve excess pressure that is required to be raised by the fan to
minimum cost, available space, available time and maintain the required flow. Pressure drop is given by:
knowledge. It is almost always desired to obtain the largest P LV 2
size of the test section and the highest speed for the available =f
funds. High speed and large size are of course competing 2Dh
demands. Aero acoustic evaluations and experiments have Where Dh is known as the Hydraulic Diameter and is
become of great importance in low-speed aerodynamics for a computed by,
number of applications. The requirement for a facility to be A
useful Aero acoustic facility is the background noise level Dh = 2
should be sufficiently low. However, now there are many
tunnels that have been modified for various degrees to is the perimeter, A is the Area of cross-section of the
reduce the background noise. It is also the case that more flow section. The term f is the friction factor and is
sophisticated measurement techniques and instrumentation computed by (Colebrook Equation):
are making it possible to obtain useful acoustic
1 2.51
measurements in environments with higher background = 2.0 log D +
noise. f 3.7 R e f
The basic considerations are: Where is the average height of surface roughness.
i) Unit Reynolds Number Since the Colebrooks equation is implicit in f f , the
determination of the friction factor requires iteration. Since
The central issue in the sizing of a low- speed wind tunnel the friction factor is minimum for smooth pipes but not
will be the achievable Reynolds number for the models that equal to zero (due to the no slip condition) and increases
can be accommodated. We will choose a Mach number of with roughness, the Colebrook Equation reduces to Prandtl
0.3 as the upper limit of Mach number for the free stream Equation and is expressed as,
beyond which we may be concerned about the effects of
Mach number on a typical study. Considering sea level f = [2log10 R e f 0.8]2
standard atmospheric conditions, the maximum for V will be This equation is valid for turbulent flows where
~100 m/sec, and the unit Reynolds number will be ~6.49 4000 R e 108 . Another general principle used to
105 m-1 (Note that the unit for Reynolds number represents calculate velocities of flow in various sections is the
Journal of Mechanical Engineering and Automation 2015, 5(3B): 47-54 49
principle of continuity of flow. their origin. The primary parameters for a diffuser are the
Q = A1 V1 = A2 V2 = C equivalent conical expansion angle and the area ratio.
Consider a conical diffuser with a radius R 1 at the entrance,
2.2. Test Section radius R 2 at the exit, and L as indicated in figure.
Commonly used cross sections are rectangular or nearly
The test sections used include shapes like round, elliptical, rectangular with reasons closely propelling the reasons given
square, rectangular, hexagonal, octagonal, and rectangular for choosing test sections with the plain walls. The
with filleted corners, flat ceiling and floor with half round equivalent conical angle is defined as follows. Let R1 be one
ends. The cost and power are directly determined by the half the entrance hydraulic diameter D1 . Let R 2 be one-half
cross-sectional area and length of the sections. The shape of A
the exit diameter D2 and AR be 2 . Then the equivalent
this test section should be based on the utility and A1
If the geometric area is constant, the speed increases in turn Diffusers are common elements in fluid flow devices, and
produces a drop in local static pressure, tending to draw the in any applications the equivalent cone angle can be quite
model downstream. In many tunnels fillets are carried large. Relatively large angles are possible for the case of
through the test section to prevent boundary layer growth in uniform flow entering the upstream end of the diffuser. If a
the corners of the test section and down the diffuser tapering very long 2.50 diffuser is used to obtain a large contraction
out at the end of the diffuser. The length of the test section in ratio, there is a danger of separation. Therefore, the total
tunnels designed with aeronautical work as the primary focus tends to be limited to area ratios of 5:1 or 6:1, half of the area
vary from one to two times the major dimension of the ratio in each half of the diffuser. The expansion loss is more
section. The power losses in the test section are sizable, as difficult and is obtained presently only by experimental
well be seen shortly owing to the high speed; thus power can correlation. The expansion loss coefficient is represented as
be saved by keeping it short. However, contractions do not a product of two factors. One factor is a function of the
deliver a uniform velocity distribution to the beginning of the equivalent conical angle and the other is a function of
test section. Therefore a constant- area duct before the test diffuser area ratio. The result is
section is usually employed. A practical detail in the test AR 1 2
section design is the installation of sufficient windows for K ex = K e ( )
AR
viewing the model. In the course of testing it will become
necessary to see all parts of the model: top, sides, bottom and The factor K e depends significantly on the diffuser cross
as much of the front as is reasonably practical. sectional shape. The most efficient divergence is about 5.
chamber length plus the nozzle plus the settling length of the angle of attack and an empty test section the fan efficiency is
exit section. close to its maximum value of 75%. The blades do not enter
the stall region even at considerably higher loads, making
2.5. Flow Straightener sure that the wind-tunnel efficiency and flow quality stays
The stagnation chamber located just before the contraction intact during the operation of the most common kinds of
cone and is the part of the wind tunnel circuit that has the experiments. The use of axial fans can, however, create some
largest cross section area. This part has a major influence on flow quality problems, if they are subjected to very high
the flow quality in the test section. Because of the high local loads, but even with more moderate loads they can create a
pressure drop generated by flow improving devices it is low frequency pulsating variation of the stream wise flow
important to position them in the location with the lowest component. In the present tunnel this is essentially, but not
flow speed and thereby minimize their contribution to the completely, avoided by a relatively low fan blade load.
total pressure drop of the wind-tunnel circuit.
3. Wind Tunnel Construction
The structural loading on the various sections of a low
speed wind tunnel are usually less critical than the strength
needed to avoid vibrations, a significant exception being the
assurance that the drive motor will stay in place should it lose
one-half of its blades. The rest of the tunnel may be
Figure 2. Flow straightener a) circular; b) square; c) hexagonal
examined to withstand the maximum stagnation pressure
with a safety factor of perhaps 4.0. Since vibration of parts of
Flow straightener is elements used to smoothen and the wind tunnel contributes to noise, discomfort of the tunnel
straighten the turbulent and irregular flow that is sucked in crew and possible fatigue failures and usually adds to the
by the wind tunnel. They are usually straightening sections turbulence in the wind stream, it is a goof practice to have the
which are squares, rectangles hexagons and circles. The natural frequencies of all tunnel part well above any exciting
extend up to 6 inches or so including screens separated by frequencies. At this time, for the small tunnels at least, flat
certain distances which make the streamline even smoother panels can be checked with a simple shaker motor, and by
and finer. The screens are very effective in breaking up means of a vibrometer the natural frequencies can be
larger eddies and acts primarily to reduce mean determined. Any speed below the fan rpm should be
non-uniformities and fluctuations of the stream wise increased by stiffening the part.
component. The screens also reduce cross flow components All types of materials are used for tunnel construction:
but less effectively than honeycombs or squares. Hence, the wood, plywood, thin metal, heavy metal, cast concrete,
combination of a straightener and screens provide, in a granite and plastics. Even though there is general
natural way, a good overall flow quality improvement. We agreement that low speed wind tunnels require no cooling,
have utilized wood plies and cut them into 36 x 3 sections heavy running in the summertime will usually make the
and further cut slots of 2 cm. About 52 strips like that and we tunnel operators wish they had some. Cooler outside air can
locked cum glued them to make the straightener structure be blown into the tunnel and some large tunnels spray water
and fitted it into the stagnation zone with rubber adhesive. over the outside to cool the tunnel. Fan blades for low speed
DRIVE SYSTEM: The fan is located downstream the tunnels are frequently made of wood, although modified
diffuser. The fan is positioned in a circular casing made of with aircraft propellers are sometimes used with trailing edge
MS sheet. Some expansion is also taking place in the flaps to provide a uniform pressure rise across the fan disk.
wind-tunnel with expanding diffuser place, located in the Tunnels that use short span, large chord blades have used
silencers to keep the total wind-tunnel circuit length to a fiberglass and more recently composite materials such as
minimum. In the center of the silencers, cylinder shaped carbon fiber reinforced epoxies.
bodies made of perforated steel plate and filled with acoustic For small research and instruction tunnels plywood is
insulation material are mounted. In the upstream silencer the possibly the best material. Particle board is another material
cylinder has a constant diameter and its nose is shaped as half used: however, it is not stiff as plywood and can develop a
an Ellipsoid. The central body in the downstream silencer is sag when supported. A smooth surface finish is difficult to
shaped as a cone to avoid flow separation. The central bodies patch. With wood holes and discontinuities can easily be
improve the flow quality both upstream and downstream of patched. The tunnel is often made in sections that bolt
the fan. They also improve the noise reduction capability of together. To prevent leaks, surgical tubing can be used as a
the silencers. The motor delivers 2HP of power, being a 3 gasket. The fan area can be metal, as it is easy to roll up
phase induction type rated for 2880rpm. It is mounted behind round sections. A door in the floor in the plane of the fan will
the fan coupled through pulleys stepping down the speed to be necessary so that blades may be taken out of their hubs
1960rpm.The fan diameter is 24 and the hub diameter is 4. and replaced.
The fan has six blades with an angle of attack of 30. With this
Journal of Mechanical Engineering and Automation 2015, 5(3B): 47-54 51
4412 aerofoil was tested for constant wind speed of 35m/s specifications of the aerofoil in the expression,
and varying the angles of attack, following observations 1
were recorded. It is visible that coefficient of lift increases F l = Cl AV 2
for positive angles of attack and becomes maximum at 35o. 2
This gives a clear indication coefficient of lift (Cl), becomes We will be able to calculate either of the values as per
maximum at 35, and on further increase of angle of attack, required for suitable Reynolds number. In order to scale this
starts to decrease and approaches a negative domain after 45. aerofoil model to that of the actual size of an Ultra-light low
At this state the aerofoil experiences zero lift and attains powered plane, we substitute the value of coefficient of lift
stall condition. An aircraft experiencing this state will fall and enter the values of density, area of cross section and a
freely due to gravity. Drag coefficient on the other hand suitable wind speed, keeping the proper limits of Reynolds
decreases with the increase of angles of attack until a point Number within standards, the actual lift for the aircraft can
where the aerofoil is on the verge of attaining stall be determined. It works the same way for drag forces as well.
condition, and then slowly increases for higher tilts of the This is the specialty of Similarities and the use of
aerofoil. dimensional analysis. The temperature sensor gave us linear
In aerodynamics, the lift-to-drag ratio, or L/D ratio, is the outputs with small variations above and below the room
amount of lift generated by a wing or vehicle, divided by temperature proving the existence of a small static
the drag it creates by moving through the air. A higher or temperature change at different stages while testing an
more favorable L/D ratio is typically one of the major goals aerofoil. The average noise value obtained from inside the
in aircraft design; since a particular aircraft's required lift is test section considering full load gave us a reading of 93dB
set by its weight, delivering that lift with lower drag leads which is a safe level as per wind tunnels are considered.
directly to better fuel economy, climb performance, and Better noise reduction would have been possible if the tunnel
glide ratio. calculating speed and estimating suitable weight sections where made circular. The designed and tested
of the aircraft we can clearly understand the correlation of lift aerofoil was the NACA 4412. Based on our results, we can
and drag forces. The values of coefficients of lift and drag conclude that the aerofoil is an excellent choice for low
obtained play a crucial role in scaling aircraft models. The powered aircrafts usually mono seat or dual seat. The
NACA 4412 Aerofoil had 20cms chord length and 5 thick. aerofoil has good lift ratios at low velocities ranging between
This aerofoil was affected by a range of lift and drag forces 30m/s to 45m/s and improves at higher velocities. The
for a wind speed of 35m/s, as shown in table. Considering aerofoil has its highest Lift to drag ratio at an angle of attack
any value of Coefficient of lift and entering the of 60.
0.2 0.16
0.15 0.14
0.1
0.12
0.05
0.1
0
Cd
-32-24-16 -8 0 8 16 24 32 40 48 56 0.08
-0.05
Cl
-0.1 Cl 0.06 Cd
-0.15 0.04
-0.2 0.02
-0.25
0
-0.3
-32-24-16 -8 0 8 16 24 32 40 48 56
-0.35
Angle of Attack ( deg) Angle of Attack ( deg)
(a) (b)
Figure 5. Variation of a) Co-efficient of Lift; b) Co-efficient of Drag with respect to angles of attack
Journal of Mechanical Engineering and Automation 2015, 5(3B): 47-54 53
10 10
8 8
6 6
4 4
2 2
0 0 LIFT
-2 0 10 20 30 40 50 60
-30 -20 -10 -32-24-16-2
-8 0 8 16 24 32 40 48 56 DRAG
L/D
-4 -4 Cd
L/D Cl
-6 -6
L/D
-8 -8
-10 -10
-12 -12
-14 -14
-16 -16
-18 -18
Angle of Attack ( deg) Angle of Attack ( deg)
(c) (d)
Figure 6. Variation of c) Length to Daimeter (L/D) ratio; d) Combination of all aerodynamic parameters with respect to angles of attack
Table 1. Experimental values for Co-efficient of Lift and Drag for different angles of attack
from Texas Instruments was chosen. A simple low-pass filter [3] John Rajadas, Bradley Rogers, Design, Fabrication and
was used with a moderate capacitance to output stable values. Testing of a Low speed Wind Tunnel Laboratory, AC
2007-1191.
Since immediate response was not a requirement, the
capacitance value was of the utmost importance. The [4] Bradshaw P. and Mehta R. D., Design Rules for Small Wind
working of this amplification and conditioning circuit was Tunnels, Aeronautical Journal, November 1979, p.p.
evident in the smooth data output values. The LM324 IC was 443-449.
used for noise and temperature output for its good stability [5] Frank, P. B., Handbook Selection, Application and Design,
for dynamic data. The results obtained from the aerofoil are McGraw-Hill, 1997.
satisfactory with results co-relating to Bernoullis theory and
[6] Milan Vilaginac, Design, Construction and Evaluation of a
general wing principles with maximum Lift to drag ratio in Subsonic Wind Tunnel, M.Sc. thesis, MIT, June 1970.
excess of 8 at 6.
[7] Pop a. and Harper j. j., Low Speed Wind Tunnel Testing,
McGraw-Hill, 1986.
[8] R. Allan Wallis, Axial Flow Fan and Ducts, John Wiley and
sons, 1983.
REFERENCES
[9] Raikan S. Dawas, Proposal for Subsonic Wind Tunnel
[1] Jewel B. Barlow, William H. Rae Jr, Alan Pope, Low speed Suitable in Iraq, M.Sc. Thesis, University of Baghdad,
Wind Tunnel Testing, 3rd edition, Wiley-India, 2010. September, 2001.
[2] Metha, R. D., The Aerodynamic Design of Blower Tunnels
with Wide Angle Diffusers, Prog. Aerospace Sci., 18, 59-120,
1977.